LGTM1 with questions: - why was the 4ms threshold chosen? - can you please send the TAG an FYI? - presumably this works with all images, CORS'd or not?
Thanks On Wednesday, November 20, 2024 at 7:41:17 AM UTC-8 Noam Rosenthal wrote: > On Wed, Nov 20, 2024 at 3:35 PM Yoav Weiss (@Shopify) < > yoavwe...@chromium.org> wrote: > >> >> >> On Wednesday, November 13, 2024 at 1:14:43 PM UTC+1 Chromestatus wrote: >> >> Contact emails nrosent...@chromium.org, mmo...@chromium.org >> >> Explainer https://github.com/w3c/paint-timing/blob/main/presentation- >> timestamps.md#security--privacy-self-review >> >> Specification https://w3c.github.io/paint-timing/#mark-paint-timing >> >> Design docs >> https://docs.google.com/document/d/1VxgMf1wlWzB4ViAW4ohkOe3AT0wQZ >> Kk7hC3IVq-cuw0/edit?tab=t.0#heading=h.fmic3y1ir4 >> >> Summary >> >> All element-timing and LCP performance entries would have a non-zero >> renderTime, even if they are cross-origin without Timing-Allow-Origin. All >> presentation timestamps (renderTime, paint timing start time, event timing >> end time) will be coarsened to a 4ms multiple to mitigate the risk of >> reading cross-origin image information. >> >> >> Blink component Blink>PerformanceAPIs >> <https://bugs.chromium.org/p/chromium/issues/list?q=component:Blink%3EPerformanceAPIs> >> >> >> TAG review None >> >> >> I agree that given the lack of API shape changes, no TAG review is >> required. >> >> >> >> >> TAG review status Not applicable >> >> Risks >> >> >> Interoperability and Compatibility >> >> This would un-zero metrics in RUM dashboards, which is generally >> positive. The coarsening might move LCP/FCP metrics by ~2ms in average, >> which RUM providers should be notified on. >> >> Have y'all done some outreach? >> > Yes, we've discussed this with RUM providers in 1:1, in social media > posts, Slack, and also on the WebPerfWG. > Will reach out again close to shipping time. > >> >> >> >> Will this feature be supported on all six Blink platforms (Windows, Mac, >> Linux, ChromeOS, Android, and Android WebView)? No >> >> >> I'm guessing this should be "yes"? >> >> > Oops, yes. updated the chromestatus entry. > > -- You received this message because you are subscribed to the Google Groups "blink-dev" group. To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+unsubscr...@chromium.org. To view this discussion visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/8f8e7a2e-1b3c-4fbd-9f42-70bc643ce5f4n%40chromium.org.
